Stock Selection Strategy: Spring-Run Chinook Salmon

Transcription

1 Spring-Run Chinook Salmon November 2010

2

3 Table of Contents 1.0 Introduction Stock Selection Strategy Development Process Donor Stock Selection Risks and Uncertainties Stock Descriptions Feather River Historic Conditions Existing Conditions Life History/Phenotypic Expression Population Size Hatchery Influence and Interbasin Transfers Deer and Mill Creeks Existing Conditions Life History/Phenotypic Expression Population Size Hatchery Influence and Interbasin Transfers Butte Creek Introduction Existing Conditions Life History/Phenotypic Expression Population Size Hatchery Influence Other Central Valley Phenotypic Spring-Run Chinook Salmon Populations Existing Conditions Life History/Phenotypic Expression Existing Population Size Hatchery Influence and Interbasin Transfers Genetics Population Genetics Lower San Joaquin River Existing Conditions Spring-Run Chinook Salmon i November 2010

4 San Joaquin River Restoration Program 6.0 Stock Comparison Population Census Life History/Phenotypic Characteristics Environmental Conditions Population Genetics Assessment and Prediction of Stock Success for Restoration Feather River Deer and Mill Creeks Butte Creek Recommendations Preferred Recommendation References Tables Table 3-1. Feather River Fish Hatchery Temperature Objectives Table 3-2. Annual Escapement Estimates for Deer Creek Table 3-3. Annual Escapement Estimates for Mill Creek Table 3-4. Butte Creek SRCS Spawning Escapement Estimates for the Period 1954 Through Table 6-1. Population Census Size from the Three Candidate Stocks Table 6-2. General Life History Characteristics for the Three Candidate Stocks Table 6-3. Population Census Size from the Three Candidate Stocks Table 6-4. Genetic Characteristic Comparison ii November 2010 Spring-Run Chinook Salmon

5 Table of Contents Figures Figure 3-1. Feather River Low-Flow and High-Flow Channel System Figure 3-2. Mean Monthly Flows in the Feather River for the Pre-Oroville Dam and Post-Oroville Dam Period Figure 3-3. Central Valley Spring-Run Chinook Salmon Spawning Run Size Composition Figure 3-4. Spring-Run Chinook Salmon Holding and Spawning Habitat in Deer Creek Figure 3-5. Spring-Run Chinook Salmon Holding and Spawning Habitat in Mill Creek Figure 3-6. Reaches of Butte Creek and West Branch of the Feather River Controlled by Pacific Gas & Electric Company Affecting Butte Creek Spring-Run Chinook Salmon, Including Temperature and Flow Gage Locations and Distances Figure 3-7. Mean Daily Water Temperature at Quartz Bowl Pool for Period July Through September Figure 3-8. Distribution by Reach of the Number of Butte Creek SRCS Holding, During Figure 3-9. Distribution by Reach of the Number of Butte Creek SRCS Spawning, During Figure Mokelumne River Figure Stanislaus River Figure 5-1. San Joaquin River Restoration Area and the Defined River Reaches Figure 6-1. Lower Elevation Water Temperature for Butte, Mill and Deer Creeks, Feather River, and Feather River Hatchery Figure 6-2. Higher Elevation Water Temperature for Butte, Mill, and Deer Creeks, Feather River, and Feather River Hatchery Figure 6-3. Taken from Framework for Assessing Viability of Threatened and Endangered Chinook Salmon and Steelhead in the Sacramento- San Joaquin River Basin Spring-Run Chinook Salmon iii November 2010

6 San Joaquin River Restoration Program List of Abbreviations and Acronyms C degrees Celsius F degrees Fahrenheit cfs centimeters per second cm centimeter Columnaris Flavobacterium columnare CVPIA Central Valley Project Improvement Act CWT coded wire tag Delta Sacramento-San Joaquin Delta DFG California Department of Fish and Game DO dissolved oxygen DWR California Department of Water Resources ESU evolutionarily significant unit FERC Federal Energy Regulatory Commission FL fork length FMP Fisheries Management Plan FMWG Fisheries Management Work Group GAPS Genetic Analysis of Pacific Salmonids GSI genetic stock identification HFC high-flow channel Ich Ichthyophthirius multiphilis JSA Joint Settlement Agreement km kilometer LFC low-flow channel LVNP Lassen Volcanic National Park m meter mm millimeter Ne effective population size NMFS National Marine Fisheries Service NOAA National Oceanic and Atmospheric Administration NRDC National Resources Defense Council PBT parentage-based tagging PG&E Pacific Gas & Electric Company PMT Program Management Team Reclamation U.S. Department of the Interior, Bureau of Reclamation iv November 2010 Spring-Run Chinook Salmon

7 Table of Contents Restoration Area RKM RM Settlement SJRRP SNP USFWS San Joaquin River between Friant Dam and the confluence with the Merced River river kilometer river mile San Joaquin River Settlement San Joaquin River Restoration Program single nucleotide polymorphism U.S. Fish and Wildlife Service Spring-Run Chinook Salmon v November 2010

8 San Joaquin River Restoration Program This page left blank intentionally. vi November 2010 Spring-Run Chinook Salmon

9 1.0 Introduction This document is part of a multi-step process to select a stock or stocks of spring-run Chinook salmon for reintroduction to the San Joaquin River and ultimately determine appropriate methods of reintroduction. The effort is part of the San Joaquin River Restoration Program (SJRRP), whose charge is to execute a legal settlement from the lawsuit, NRDC et al. v. Kirk Rodgers et al.; whereby in 1988, a coalition of environmental groups, led by the Natural Resources Defense Council (NRDC), filed a lawsuit challenging the renewal of long-term water service contracts between the United States and California s Central Valley Project Friant Division contractors. After more than 18 years of litigation, the Settling Parties reached a Stipulation of Settlement Agreement (Settlement). The Settling Parties, including NRDC, Friant Water Users Authority, and the U.S. Departments of the Interior and Commerce, agreed on the terms and conditions of the Settlement, which was subsequently approved on October 23, The Settlement establishes two primary goals: Restoration Goal To restore and maintain fish populations in good condition in the mainstem San Joaquin River below Friant Dam to the confluence with the Merced River, including naturally reproducing and self-sustaining populations of salmon and other fish. Water Management Goal To reduce or avoid adverse water supply impacts to all of the Friant Division long-term contractors that may result from the Interim Flows and Restoration Flows provided for in the Settlement. Related to the Settlement, President Obama signed the San Joaquin River Restoration Act on March 30, 2009, giving the U.S. Department of Interior full authority to implement the SJRRP. The implementing agencies, consisting of the U.S. Department of Interior, Bureau of Reclamation (Reclamation) and U.S. Fish and Wildlife Service (USFWS), National Marine Fisheries Service (NMFS), California Department of Fish and Game (DFG), and California Department of Water Resources (DWR) organized a Program Management Team (PMT) and associated Work Groups to begin work implementing the Settlement. The Fisheries Management Work Group (FMWG), consisting of representatives of the above agencies, prepared the Fisheries Management Plan (FMP) to describe the program s approach to restoration. This Stock Selection Strategy works to fulfill the stock selection objectives of the FMP with focus on the three largest stocks of spring-run Chinook salmon in the Central Valley: Feather River, Butte Creek, and the Deer and Mill Creek Complex. A general description of each stock and their river system is provided, as well as an analysis and comparison of each stock s genotypic and phenotypic characteristics and recommendations for stock selection. Spring-Run Chinook Salmon 1-1 November 2010

10 San Joaquin River Restoration Program 1.1 Stock Selection Strategy Development Process This document is the product of the Genetics Subgroup of the FMWG. The Genetics Subgroup focuses on genetic issues related to protecting the genetic integrity of the reintroduced stock, stock selection, reintroduction strategies, development of the Hatchery and Genetics Management Plan, and other hatchery-related issues. This subgroup is composed of State and Federal fisheries scientists and academic researchers. This document is guided by an adaptive management approach, as described in the FMP. While extensive analysis and expertise is used to predict stock performance in the restored environment, it is recognized that these predictions are potentially fallible due to the numerous variables associated with the massive scale of this project. A key aspect to this decision-making process is the use of adaptive management, as described by Williams et al. (2009), which recognizes and embraces this uncertainty. Making a sequence of good management decisions is more difficult in the presence of uncertainty, an inherent and pervasive feature of managing ecological systems (16, 17). Uncertainties arise with incomplete control of management actions, sampling errors, environmental variability, and an incomplete understanding of system dynamics, each affecting the decision making process. An adaptive approach provides a framework for making good decisions in the face of critical uncertainties, and a formal process for reducing uncertainties so that management performance can be improved over time. For more information about the adaptive management process use here, refer to Chapter 1 of the FMP. 1-2 November 2010 Spring-Run Chinook Salmon

11 2.0 Donor Stock Selection Spring-run Chinook salmon once occupied all major river systems in California where there was access to cool reaches that would support over-summering adults. Historically, spring-run Chinook salmon were widely distributed in streams of the Sacramento-San Joaquin river basins, spawning and rearing over extensive areas in the upper and middle reaches (elevations ranging 1,400 to 5,200 feet (450 to 1,600 meters (m))) of the San Joaquin, American, Yuba, Feather, Sacramento, McCloud, and Pit rivers (Meyers et al. 1998). Spring-run Chinook salmon populations in the San Joaquin River Basin were extirpated following basin-wide dam construction between 1894 to 1968 (Yoshiyama et al. 2001, Lindley et al. 2004, Schick and Lindley 2007) and all extant spring-run Chinook salmon populations are believed to spawn in the Sacramento River Basin (Moyle 2002). In the upper San Joaquin River, spring-run Chinook salmon were extirpated by the mid-to late 1940s, following the construction of Friant Dam and diversion of water for agricultural and municipal purposes (e.g., Central Valley Project) to the San Joaquin Valley. Only two evolutionarily significant units (ESU) of spring-run Chinook salmon remain in California: the Central Valley spring-run Chinook salmon ESU, consisting of four Central Valley spring-run Chinook salmon populations, and the Upper Klamath-Trinity Rivers Chinook salmon ESU, which includes all naturally spawning spring-run Chinook salmon in the Klamath and Trinity basins upstream from the confluence of the Klamath and the Trinity Rivers (Moyle et al. 1995). Only Chinook salmon from the Central Valley ESU will be considered for reintroduction. Lindley et al. (2004) used ecogeomorphic principles to identify at least 18 historic spring-run Chinook salmon populations in the Sacramento-San Joaquin watershed. While the genetic constituency of these historic populations is uncertain, it is possible that each population was sufficiently isolated and maintained some level of genetic distinctiveness in the face of limited gene flow. Functionally independent populations of spring-run Chinook salmon remain in Deer, Mill, and Butte creeks and another spring-run Chinook salmon population is spawned at the Feather River Fish Hatchery (FRFH) and in the river below Oroville Dam. Spring-run Chinook salmon also occur in numerous smaller northern Central Valley tributaries, though these populations are small and subject to gene flow from the larger independent populations in the California Central Valley. Several tributaries within the San Joaquin Valley have spring-run Chinook salmon, but their numbers are very small and further monitoring and research is needed to determine if these fish are genotypically spring-run Chinook salmon, or fall-run Chinook salmon. Spring-run Chinook salmon populations are phenotypically similar in their adult behavior patterns. They return to natal rivers sexually immature in the spring, typically ascending farther upstream than later-entering fall-run Chinook salmon, then reside in cool water refugia until spawning starts early in the fall. Life history differences among spring-run Chinook salmon populations are informative in considering their potential use in Spring-Run Chinook Salmon 2-1 November 2010

12 San Joaquin River Restoration Program reintroduction actions and it is possible this phenology and local adaptation have led to underlying genetic differences among these groups. Research in other salmonids have described the local adaptation of egg incubation temperature optima, yolk conversion efficiencies, development rates, subyearling growth rates, and age at smoltification (Hendry et al. 1998, Obedzinski and Letcher 2004), which may cause differential survival among stocks in environments distinct from the natal streams. Reintroduction efforts may have the best chance for success when the chosen broodstock have life history characteristics compatible with the anticipated environmental conditions of the reintroduction habitat. Ecoregions closest to the restoration site that contain Chinook salmon populations have the highest likelihood of similar local adaptation of traits and, therefore, only Chinook salmon populations found in California s Central Valley will presently be considered as broodstock. The primary goal of broodstock selection is to identify the stock(s) with the highest likelihood of establishing a self-sustaining, naturally reproducing population in the San Joaquin River Restoration Area (San Joaquin River between Friant Dam and the confluence with the Merced River). A key component to identifying the best stock(s) is conducting genetic analyses of extant populations to ascertain the genetic integrity of all potential source populations. Measurement indices that are useful for analysis of potential broodstock(s) include, but are not limited to: effective population size (Ne); genetic comparisons to historic population in upper San Joaquin (if feasible); within population genetic diversity and inbreeding levels; among population genetic diversity; and hatchery influence. Optimum characteristics for the chosen donor population sources include: Be of local or regional origin (Central Valley) Have life history (behavioral and physiological) characteristics that fit conditions expected to occur on the San Joaquin River, thereby maximizing the probability of successful reintroduction Large effective population size High within population genetic diversity with low inbreeding coefficients Adequate representation of overall ESU genetic diversity The candidate populations for this program may be limited to those with relatively large effective population size; the independent spring-run Chinook salmon populations on Deer/Mill and Butte creeks, and spring-run Chinook salmon population in the Feather River. All potential sources of spring-run Chinook salmon are analyzed in this document. In addition to genetic considerations, the appropriate broodstock(s) for the project will be selected based on current (census) population size, compatibility of life history characteristics to anticipated restored Restoration Area conditions, and availability of broodstock. This information will be gathered through interactions with biologists for all potential source populations and review of existing literature and databases. 2-2 November 2010 Spring-Run Chinook Salmon

13 2.0 Donor Stock Selection 2.1 Risks and Uncertainties Selected broodstock(s) will not capture the genetic variation needed to promote a long-term naturally self-sustaining population in the Restoration Area. An assessment of each potential broodstock s genetic diversity (e.g., Ne, heterozygosity) is proposed to ensure that the chosen source population(s) possesses adequate variation to adapt to changing environmental conditions. Genetic analyses will be facilitated by genotyping a large number of single nucleotide polymorphism (SNP) markers. Selection of multiple broodstocks could act to reduce risk by increasing overall genetic variation. An overlap in migration run-timing and lack of spatial separation between mature spring-run Chinook salmon and fall-run Chinook salmon in the Restoration Area are expected to result in the genetic introgression of the two populations. To reduce the potential for hybridization, it is recommended that a physical barrier (e.g., weir) be installed after the spring-run Chinook salmon spawning migration is completed to separate upstream spring-run Chinook salmon spawning habitat from the downstream fall-run Chinook salmon spawning habitat. Due to overlap in spring-run Chinook salmon and fall-run Chinook salmon spawning migrations, reestablishment of late fall-run Chinook salmon may be preferable over early fall-run Chinook salmon spawners. Removal of broodstock fishes from source population(s) may increase the risk of extirpation, and reduce the population viability and recovery potential of the source population(s). To reduce the potential for significant impacts to source population(s), criteria for collection strategies will balance development of reintroduced stocks with minimizing risks to the source population(s). Spring-Run Chinook Salmon 2-3 November 2010

14 San Joaquin River Restoration Program This page left blank intentionally. 2-4 November 2010 Spring-Run Chinook Salmon

15 3.0 Stock Descriptions 3.1 Feather River The Feather River is a major tributary to the Sacramento River located at the northern end of the western slope of the Sierra Nevada, with a watershed encompassing 5,900 square miles (FERC 2007, NMFS 2009). The upper Feather River watershed above Oroville Dam, is approximately 3,600 square miles (approximately 68 percent of the Feather River Basin), and has four tributaries, the North, South, Middle and West forks. Downstream from Oroville Dam, the watershed includes the drainage of the Yuba and Bear rivers, and eventually meets the Sacramento River, contributing 25 percent to its flow (NMFS 2009) Historic Conditions The Feather River is renowned as one of the major salmon-producing streams of the Sacramento Valley (Yoshiyama et al. 2001) and once contained more than 200 miles of anadromous fish habitat, of which 64 miles remain (NMFS 2009). Before the construction of numerous hydroelectric power projects and diversions, spring-run Chinook salmon ascended high into the watershed (Clark 1929, Yoshiyama et al. 1996, Lindley et al. 2004). The fall-run Chinook salmon spawned primarily in the mainstem, while most of the spring-run Chinook salmon spawned in the Middle Fork, with smaller runs in the North, South and West Forks (Fry 1961, Yoshiyama et al. 2001). Each of the four tributaries above Oroville Dam generally provide suitable habitat for all life stages of Chinook salmon and steelhead (DWR 2005, NMFS 2009) and likely contained independent populations of spring-run Chinook salmon (Lindley et al. 2004). Human impacts to the salmon runs of the Feather River began as early as the late 1800s. Hydraulic mining activity and dam construction, where established below Oroville and on the West, North, and South forks, occurred in the early 1900s (Clark 1929, Muir 1938, as found in Yoshiyama et al. 2001); up to 186 million cubic yards of mining debris were produced before 1909 (Gilbert 1917, Yoshiyama et al. 2001). Fry (1961) reported run-size estimates for the fall-run Chinook salmon of 10,000 to 86,000 fish during the period 1940 to 1959, and about 1,000 to 4,000 spring-run Chinook salmon. Just before the completion of Oroville Dam, a small naturally spawning springrun Chinook salmon population still existed in the Feather River (Reynolds et al. 1993, Yoshiyama et al. 2001). The number of naturally spawning spring-run Chinook salmon in the Feather River was estimated only periodically in the 1960s and 1970s, with estimates ranging from 2,908 fish in 1964 to two fish in 1978 (NMFS 2009). Spring-Run Chinook Salmon 3-1 November 2010

16 San Joaquin River Restoration Program Existing Conditions Flow Regime Today, flow in the Feather River is altered by hydroelectric, water storage, and diversion projects (FERC 2007). River flow below the reservoir is regulated by Oroville Dam, Thermalito Diversion Dam, and the Thermalito Afterbay Outlet. Oroville Reservoir is the lowermost reservoir on the Feather River and the upstream limit for anadromous fish (USFWS 1995, NMFS 2009). Under normal operations, the majority of the Feather River is diverted at Thermalito Diversion Dam into Thermalito Forebay. The remainder of the flow, typically 600 cubic feet per second (cfs), flows through the historical river channel, referred to as the lowflow channel (LFC) (Figure 3-1). Mean monthly flows through the LFC are now significantly less than pre-dam levels (Sommer et al. 2001) (Figure 3-2). Water released by the Thermalito Forebay is used to generate power before discharge into the Thermalito Afterbay and enters the high-flow channel (HFC), then water flows southward through the valley until the confluence with the Sacramento River at Verona (FERC 2007). 3-2 November 2010 Spring-Run Chinook Salmon

17 3.0 Stock Descriptions Source: DWR Figure 3-1. Feather River Low-Flow and High-Flow Channel System Spring-Run Chinook Salmon 3-3 November 2010

18 San Joaquin River Restoration Program Source: DWR Note: Total flow in the post-dam period includes the portion from the low channel and the portion diverted through the Thermalito Complex. Figure 3-2. Mean Monthly Flows in the Feather River for the Pre-Oroville Dam ( ) and Post-Oroville Dam ( ) Period Geology The North Fork Feather River is in the southern Cascades while the other forks are in the Sierra Nevada ecoregion. The headwaters of the North Fork are fed by rainfall and by snowmelt from Mt. Lassen, and rocks are predominately of volcanic origin (Lindley et al. 2004). The bed material in the remaining three tributaries is primarily of granitic origin. As described in NMFS 2009, the most common material in the soils downstream from Oroville Dam is alluvium, with some soils derived from debris deposited during the hydraulic mining period. Channel banks and streambed in the LFC generally consist of armored cobble as a result of periodic flood flows and the absence of gravel recruitment. By far, historic hydraulic mining of gold-bearing gravel deposits has caused the largest impact on the Feather River channel, washing massive amounts of erosional debris, including cobbles, gravel, sand, silt, and clay, into the river. Floodplain soils are conducive to agriculture and many areas of riparian floodplain and fluvial terraces have been converted to irrigated crops and orchards (FERC 2007). Human activity over time has resulted in decreased vegetative cover from logging and grazing, channel clearing, levee construction, and water diversions. These activities have contributed to the increased sediment load in the Feather River watershed (FERC 2007). 3-4 November 2010 Spring-Run Chinook Salmon

19 3.0 Stock Descriptions Temperature and Water Quality Water is released from Oroville Dam through a multilevel outlet to provide appropriate water temperatures for the operation of the FRFH (Table 3-1) and to protect downstream fisheries (NMFS 2009). Water temperatures downstream from the Fish Barrier Dam vary seasonally and there is a significant temperature difference between the LFC and the HFC. In both channels, temperatures begin to warm in March and peak in July and early August. In the LFC, peak temperatures range from 61 degrees Fahrenheit ( F) (16 degrees Celsius ( C)) upstream from the FRFH to 69 F (21.5 C) upstream from the Thermalito Afterbay Outlet (FERC 2007). Cooling begins in September, with water temperatures dropping to 45 F (7 C) throughout the reach by February (FERC 2007). Compared to historical levels, mean monthly water temperatures in the LFC at Oroville are 2 to 14 F (1.1 to 7.8 C) cooler during May through October and 2 to 7 F (1.1 to 3.9 C) warmer during November through April (Sommer et al. 2001). FRFH water temperatures vary little from temperatures of river water near the hatchery (FERC 2007). Peak water temperatures in the HFC range from 71 to 77 F (22 to 25 C). River cooling begins in late August, with minimum temperatures of 44 to 45 F (6.7 to 7.2 C) reached by January or February. Releases from the Thermalito Afterbay Outlet as well as flow contributions from Honcut Creek, the Yuba River, and the Bear River influence HFC water temperatures between April and October (FERC 2007). Except during periods of high flow through the Thermalito Afterbay, which occur frequently in July and August, releases from the Thermalito Afterbay during the warm season generally raise the water temperature of the river. Honcut Creek and Bear River inflows also tend to increase Feather River temperatures downstream from their confluences during this period (FERC 2007). Flows contributed by the Yuba River tend to cool the Feather River during the warmer spring and summer months. Dissolved oxygen (DO) and ph levels in the Feather River are generally found to comply with the water quality objectives for Chinook salmon. When exceedances occur, they are considered minor (FERC 2007). Table 3-1. Feather River Fish Hatchery Temperature Objectives (±4 F between April 1 and November 30) Temperature Period ( F) April May May June June 16 August August September 52 October November 51 December - March No greater than 55 Source: DWR 2001 Key: F = degrees Fahrenheit Spring-Run Chinook Salmon 3-5 November 2010

20 San Joaquin River Restoration Program Life History/Phenotypic Expression Holding and Spawning Upstream migration of Chinook salmon is blocked by Fish Barrier Dam located 0.6 mile (1 kilometer (km)) below the Oroville Dam. Adult spring-run Chinook salmon are found holding at the Thermalito Afterbay Outlet and the Fish Barrier Dam as early as April (FERC 2007, NMFS 2009) and begin spawning in September, usually 2 to 3 weeks earlier than the fall-run Chinook salmon (Kindopp pers. comm.). Adult fall-run Chinook salmon typically return to the river to spawn during September through December, with peak returns from mid-october through early December (Sommer et al. 2001). Spring-run Chinook salmon are spawned artificially in the FRFH and also spawn naturally in the river during late September to late October (Reynolds et al. 1993, Yoshiyama et al. 2001) downstream from the Fish Barrier Dam approximately 8 miles to the Thermalito Afterbay Outlet (NMFS 2009). Fall-run Chinook salmon and steelhead are also produced by the FRFH. Approximately two-thirds of natural Chinook salmon spawning in the Feather River occurs in the LFC between the Fish Barrier Dam and the Thermalito Afterbay Outlet (NMFS 2009). Spawning occurs primarily in the riffle and glide areas, with the greatest portion crowded in the upper 3 miles of the LFC (Sommer et al. 2001). The remaining one-third of the spawning occurs between the Thermalito Afterbay Outlet and Honcut Creek (River Mile (RM) 59 to 44) (FERC 2007), where, in comparison to the LFC, there is a greater amount of available spawning areas and deeper pools (FERC 2007, NMFS 2009). This represents a marked shift in the spawning distribution of Chinook salmon since the construction of Oroville Dam and the FRFH, when less spawning activity occurred in the LFC, which has undoubtedly increased spawning densities in the LFC (Sommer et al. 2001). For both Chinook salmon and steelhead, spawning and embryo incubation is the life stage for which the smallest amount of suitable habitat is available in the upper Feather River (NMFS 2009). Rearing Some spring-run Chinook salmon juveniles hold over the summer in deep pools within the LFC 5 miles below Oroville Dam and the downstream Thermalito Afterbay Outlet (Reynolds et al. 1993, (Yoshiyama et al. 2001). The vast majority of spring-run Chinook salmon fish emigrate as fry (DWR unpublished data as found in Sommer et al.), suggesting that rearing habitat is limiting or that conditions later in the season are less suitable (Sommer et al. 2001). The primary location(s) where these fish rear is unknown; however in wetter years it appears that many young salmon rear for weeks to months in the Yolo Bypass floodplain immediately downstream from the Feather River before migrating to the estuary (Sommer et al. 2001). 3-6 November 2010 Spring-Run Chinook Salmon

21 3.0 Stock Descriptions Outmigration Fry from both runs of Chinook salmon emerge from spawning gravels as early as November (Painter et al. 1977, DWR unpublished data as found in Sommer et al. 2001) and generally rear in the river for at least several weeks. Emigration occurs from December to June, with a typical peak during the February-through-April period (Sommer et al. 2001), with 95 percent of the juvenile Chinook typically emigrating from the Oroville Facilities project area by the end of May (FERC 2007, NMFS 2009) Population Size The Central Valley spring-run Chinook spawn run size data between 1970 and 2008 is summarized in Figure 3-3. Between this period, the highest annual hatchery spring-run Chinook salmon escapement on the Feather River was 8,662, occurring in 2003 (DFG 2009). Between 1986 and 2007, the average number of spring-run Chinook salmon returning to the FRFH was 3,992, compared to an average of 12,888 spring-run Chinook salmon returning to the entire Sacramento River Basin (NMFS 2009), and an average of 1,700 fish before the construction of Oroville Dam (Reynolds et al. 1993, Yoshiyama et al. 2001). More recently, FRFH spring-run Chinook salmon escapement from 2005 through 2008 was 1,774, 2,061, 2,674, and 1,418, respectively (DFG 2009, NMFS 2009). The increase in numbers since the completion of the dam is attributed to the consistent supply of cold water to both the hatchery and the LFC and the contribution of hatchery fish (Reynolds et al. 1993, Yoshiyama et al. 2001). Spring-Run Chinook Salmon 3-7 November 2010

22 San Joaquin River Restoration Program Figure 3-3. Central Valley Spring-Run Chinook Salmon Spawning Run Size Composition ( ) Other Wild Mill, Deer and Butte Creeks FRFH Year R Source: DFG November 2010 Spring-Run Chinook Salmon

23 3.0 Stock Descriptions Hatchery Influence and Interbasin Transfers The FRFH was built by DWR to mitigate for the loss of upstream spawning habitat of salmon and steelhead due to the building of Oroville Dam (Reynolds et al. 1993, Yoshiyama et al. 2001). The FRFH began operation in 1967, and it is the only source of hatchery-produced spring-run Chinook salmon in the Central Valley (Reynolds et al. 1993, Yoshiyama et al. 2001). In the early stages of hatchery operations, FRFH staff attempted to maintain program separation of the two runs by designating the earliestarriving spawners as spring-run Chinook salmon. Unfortunately, directed and unintentional incorporation of fall-run Chinook salmon broodstock into the spring-run Chinook salmon program has led to hybridization between the two hatchery stocks over time. Brown and Greene (1994) describe coded-wire-tag studies on the progeny of hatchery fish identified as fall-run Chinook salmon and spring-run Chinook salmon and found evidence of substantial introgression (Sommer et al. 2001) due to hatchery practices and the overlapping spatial proximity of spawning in the river of the two populations. It has been reported that some proportions of the offspring of each hatchery race return as adults during the wrong period, i.e., spring-run Chinook salmon are returning during months when fall-run Chinook salmon return (Sommer et al. 2001). In an attempt to improve the life-history integrity of the spring-run Chinook salmon hatchery stock, a Settlement Agreement for Licensing of the Oroville Facilities (March 2006) includes measures to improve the short- and long-term genetic management of the FRFH spring-run Chinook salmon program, and measures to physically separate and isolate spring-run Chinook salmon from fall-run Chinook salmon (NMFS 2009). 3.2 Deer and Mill Creeks Deer and Mill creeks are eastside tributaries to the upper Sacramento River. Deer Creek enters the Sacramento River at RM 220 and Mill Creek enters at RM 230. Along with Butte Creek, they are recognized as supporting genetically distinct, self-sustaining populations of spring-run Chinook salmon, (DFG 1998, as cited in DFG 2008). Mill and Deer creeks appear genetically similar compared to the other extant spring-run Chinook salmon population in the Central Valley and likely function together demographically as a metapopulation. There is currently no hatchery program supplementing the populations on these streams. Between 1902 and 1940, the U.S. Bureau of Fisheries established a hatchery on Mill Creek near Los Molinos. During this time, fall-run Chinook salmon were spawned, with an average of 6,000,000 to 7,000,000 eggs taken annually. Juvenile salmon were reared and released in the spring. Attempts were made to spawn spring-run Chinook salmon at this site, but were prohibited by warm water temperatures during summer months. (Hanson et. al. 1940) Additionally, during salvage operations resulting from the construction of Keswick Dam between 1941 and 1946, about 13,000 adult spring-run Chinook salmon from the upper Sacramento River were introduced into Deer Creek (Cramer and Hammack 1952). According to Harvey (1997), some of these may have been winter- and/or fall-run Chinook salmon. Small numbers of fall-run and/or late fall-run Chinook salmon may also spawn annually in Deer and Mill creeks (Harvey-Arrison 2007) Spring-Run Chinook Salmon 3-9 November 2010

24 San Joaquin River Restoration Program Existing Conditions Deer Creek Deer Creek is 60 miles long and its watershed drains 200 square miles (USFWS 1995). Deer Creek originates on the northern slopes of Butte Mountain at an elevation of approximately 7,320 feet. It initially flows through meadows and dense forests and then descends rapidly through a steep rock canyon into the Sacramento Valley. Deer Creek flows for 11 miles across the Sacramento Valley floor, entering the Sacramento River at approximately a 180-foot elevation (NMFS 2009) where most of the flow is diverted. In many years, diversions at three dams deplete all of the natural flow from mid-spring to fall. Each of these diversion structures have fish passage structures and screens, so Deer Creek spring-run Chinook salmon have access to 100 percent of their historic habitat (Figure 3-4) (NMFS 2009). Source: Harvey-Arrison 2008 Figure 3-4. Spring-Run Chinook Salmon Holding and Spawning Habitat in Deer Creek 3-10 November 2010 Spring-Run Chinook Salmon

25 3.0 Stock Descriptions Mill Creek Mill Creek is a major tributary of the Sacramento River, flowing from the southern slopes of Mount Lassen and entering the Sacramento River at RM 230. The stream originates at an elevation of approximately 8,200 feet and descends to 200 feet at its confluence with the Sacramento River. Mill Creek originates from springs in Lassen Volcanic National Park (LVNP) and initially flows through meadows and dense forests. It descends rapidly through a steep canyon, flows eight miles across the Sacramento Valley floor, and its total length is approximately 58 miles to its confluence with the Sacramento River. Nearly all the mainstem habitat is used and/or available to spring-run Chinook salmon (Figure 3-5). The Mill Creek watershed encompasses 134 square miles. During the irrigation season, three dams on the lower 8 miles of the stream divert most of the natural flow, particularly during dry years. Source: Harvey-Arrison 2008 Figure 3-5. Spring-Run Chinook Salmon Holding and Spawning Habitat in Mill Creek Spring-Run Chinook Salmon 3-11 November 2010

26 San Joaquin River Restoration Program Life History/Phenotypic Expression Deer Creek Migration. Spring-run Chinook salmon have been documented migrating upstream on Deer Creek from March through early July. Migrations usually end during the peak of the irrigation season when flows are insufficient to pass adults and water temperatures begin to approach lethal limits low in the watershed. Holding and Spawning. The known range for adult spring-run Chinook salmon holding extends from Upper Falls downstream to near the confluence of Rock Creek, a distance of approximately 25 miles. The upstream limit is a natural waterfall (Upper Falls). Within this area, 30 percent of the area is represented by all pools. Of 166 total pools, 98 (or 60 percent) are holding pools (more than 6 feet deep). Because maturing adult spring-run Chinook salmon enter streams during the spring months and spend the summer holding in deep pools (before fall spawning), they are present in the stream system when temperatures are at their peak (generally July and August). In Deer Creek above the canyon mouth, Needham et al. (1943) observed salmon holding in deep pools when surface water temperatures measured 73 F (23 C). Based on adult spring-run Chinook salmon mortalities reported in lower Deer Creek (below the canyon mouth) in the 1940s, Cramer and Hammack (1952) reported temperatures greater than 81 F (27 C) were lethal to migrating salmon. The known range for adult spring-run Chinook salmon spawning extends from Upper Falls downstream to near the mouth of the canyon, a distance of approximately 30 miles. It appears that in wet years, more spawning takes place lower in the watersheds. Spawning habitat use has been known to shift between years at some sites with changes in bed composition resulting from high-flow events. Visual observations of spring-run Chinook salmon spawning in Deer Creek indicate spawning substrate is in good condition with the percent fines being low in the areas used. Deposition of fines in areas used for spawning is virtually absent year round. Emergence and Rearing. In 2007, DFG initiated bimonthly rearing surveys to assess the relative growth of known spring-run Chinook salmon juveniles with mixed-stock juveniles captured in rotary screw traps. In 2007, surveys began in January and juveniles were first detected in February. In 2008, surveys could not begin until March due to snow conditions, and juveniles were detected on the first survey of the season. Monitoring data indicate that emergence of juvenile Chinook begins in November, peaks around February, and ends in April. These data are derived from an egg-temperature model to predict emergence based on redd placement and also from direct observation of newly emerged juveniles. (Harvey-Arrison 2007) 3-12 November 2010 Spring-Run Chinook Salmon

27 3.0 Stock Descriptions Outmigration. Based on annual surveys by the DFG, outmigration of yearling springrun Chinook salmon typically occurs from October or November through March or April, depending on the year. Fry outmigration occurs from February through June, but since traps are located within the fall-run Chinook salmon spawning area, these fry migrations are a mix of fall-run and spring-run Chinook salmon progeny. In Deer and Mill creeks, many juveniles emigrate during the wet season more than a year after being spawned (Big Chico Creek Watershed Alliance 2000). Mill Creek Migration. While adult spring-run Chinook salmon have been observed migrating in Mill Creek as early as February, a 10-year study from 1953 to 1964 (DFG 1966) has documented the majority of upstream migration as occurring between mid-april and the end of June. Based on observations of spring-run Chinook salmon adults holding and/or spawning, the known range of this habitat extends a distance of approximately 48 miles from near the Little Mill Creek confluence (Harvey pers. comm. as cited in Armentrout 1998 and reported in NMFS 2009) upstream to within 0.5 mile of the LVNP boundary (personal observation of adult holding). Suitable spawning habitat on the mainstem of Mill Creek extends to near Morgan Hot Springs (approximately 3 miles downstream from LVNP), although salmon have been reported spawning in "Middle Creek", a small tributary located approximately two miles downstream from the park boundary (McFarland 1997). Holding and Spawning. There are two geographically important sections of holding habitat available on Mill Creek, Upper Mill Creek and Lower Mill Creek (Canyon). Upper Mill Creek, is defined as the upper 7.6 miles of Mill Creek between the LVNP boundary and Mill Creek campground, and Lower Mill Creek (canyon reach), is defined as the area downstream from the Mill Creek campground (Figure 3-5). In Upper Mill Creek, the availability of spring-run Chinook salmon holding habitat appears to be limited. Based on stream survey data collected in 1990, 5 percent of the area was represented by all pools. Of all 88 pools noted in 1990, none was classified as a holding pool. Downstream from the Mill Creek campground, in the Lower Mill Creek (Canyon) reach, available holding habitat is more abundant. In 1990 and 1994, a survey was conducted on more than 13 miles of approximately 20 miles of stream extending from the campground to 2 miles downstream from Black Rock. Within the surveyed segments, 13 percent of the area was represented by all pools. Of all 86 pools documented, 20 (or 23 percent of the total) were holding pools. Spring-Run Chinook Salmon 3-13 November 2010

28 San Joaquin River Restoration Program Little quantifiable data is available on the distribution of holding habitat from approximately 2 miles downstream from Big Bend to approximately 2 miles upstream from Black Rock due to the difficulty in accessing the area. In a 1988 holding survey, more than 200 adult salmon were noted within most of the 7 miles of stream that had not been previously habitat classified, indicating additional suitable holding habitat is present. Given similar channel characteristics such as substrate composition, gradient, etc., holding habitat distribution and abundance would not likely differ greatly from other areas of Mill Creek surveyed in the lower canyon reaches (McFarland 1997). Above the canyon mouth, in the upper alluvial reach of Mill Creek, is an area of possible temperature-related impacts on adults. Adult mortalities have been reported during midsummer in a single drought year (McFarland 1997). The area where the mortalities occurred contained natural hot springs and lacked deep holding pools. The stream channel was mostly open with little riparian shading and overhead cover, the mortalities may have been attributed to a prolonged exposure to elevated stream temperatures. Mill Creek spring-run Chinook salmon are unique for spawning at an elevation of more than 5,000 feet, the highest elevation known for salmon spawning in North America (Armentrout et al In Mill Creek, sediment loading is greater than in Deer Creek and fines are notable, especially in areas of deposition. High gravel embeddedness has been observed in some areas of spawning use (McFarland 1997). Conditions observed however, do not appear to limit salmon from spawning. Size distribution of Mill Creek spring-run Chinook salmon spawners has ranged from 41 cm to 102 cm from carcass survey data from 1990 to The majority are in the 60- to 80-centimeter (cm) fork length (FL) range. Emergence and Rearing. In 2007, DFG initiated bimonthly rearing surveys to assess the relative growth of known spring-run Chinook salmon juveniles with mixed stock juveniles captured in rotary screw traps. In 2007, surveys were initiated in January and juveniles were first detected in February. In 2008, surveys could not begin until March due to snow conditions, and juveniles were detected on the first survey of the season. Monitoring data indicate that emergence of juvenile Chinook begins in November, peaks around February and ends in April. These data are derived from an egg-temperature model to predict emergence based on redd placement and also from direct observation of newly emerged juveniles (Harvey-Arrison 2007). Outmigration. Based on annual surveys by the DFG, outmigration of yearling springrun Chinook salmon typically occurs from October or November through March or April depending on the year. Fry outmigration occurs from February through June, but since traps are located within the fall-run Chinook salmon spawning area, these fry migrations are a mix of fall-run and spring-run Chinook salmon progeny. In Deer and Mill creeks, many juveniles emigrate during the wet season more than a year after being spawned (Big Chico Creek Watershed Alliance 2000) November 2010 Spring-Run Chinook Salmon

29 3.0 Stock Descriptions Population Size Deer Creek Table 3-2 shows annual escapement estimates for Deer Creek spring-run Chinook salmon. For the Central Valley Project Improvement Act (CVPIA) doubling period 1967 to 1991, the average spawning escapement of spring-run Chinook salmon in Deer Creek was 1,300 (USFWS 1995). From 1991 to 2008, the average is only 1,152. Table 3-2. Annual Escapement Estimates for Deer Creek Year Count Year Count Year Count , , , , , , , , , , , , , , , , , , Source: DFG 2009 Spring-Run Chinook Salmon 3-15 November 2010

30 San Joaquin River Restoration Program Mill Creek Table 3-3 shows annual escapement estimates for Mill Creek spring-run Chinook salmon. For the CVPIA doubling period 1967 to 1991, the average spawning escapement of spring-run Chinook salmon in Mill Creek was 800 (USFWS 1995). From 1991 to 2008, the average is only 646. Table 3-3. Annual Escapement Estimates for Mill Creek Year Count Year Count Year Count , , , , , , , , , , , , , , , Source: DFG Hatchery Influence and Interbasin Transfers There is currently no hatchery program supporting fish populations on either of these streams. Between 1902 and 1940, the U.S. Bureau of Fisheries established a hatchery on Mill Creek near Los Molinos. During this time, fall-run Chinook salmon were spawned, with an average of 6,000,000 to 7,000,000 eggs taken annually. Juvenile salmon were reared and released in the spring. Attempts were made to spawn spring-run Chinook salmon at this site, but were prohibited by hatchery warm water temperatures during the summer months (Hanson et. al. 1940) November 2010 Spring-Run Chinook Salmon

31 3.0 Stock Descriptions 3.3 Butte Creek Introduction Butte Creek is one of only three streams to sustain a genetically distinct and viably independent population of spring-run Chinook salmon (NMFS 2009). The spring-run Chinook salmon in Butte Creek are considered persistent and viable and is one of the most productive spring-run Chinook salmon streams in the California Central Valley (NMFS 2009). Lindley et al. (2007) indicated that the Butte Creek population is at a low risk of extinction due to the population size, general increases in production, and low hatchery influence. According to Moyle et al. (2008), there is a high likelihood of springrun Chinook salmon going extinct in the next 50 to 100 years due to the vulnerability of a catastrophic event and due to the narrow physiological tolerances in the summer, where an increase in temperature due to climate change may drastically reduce survival. Population numbers have increased within the last 2 decades, and large pre-spawn mortalities have occurred in a few years (Williams 2006). The pre-spawn mortalities were due to a high number of fish concentrated in limited holding pools with high water temperatures, resulting in an outbreak of diseases Existing Conditions Flow Regime The flow regime of the adult holding and spawning habitat in Butte Creek is directly affected by the Pacific Gas & Electric (PG&E) DeSabla-Centerville Project (Figure 3-6) (FERC-083). The entire holding and spawning habitat for spring-run Chinook salmon is located downstream from the Centerville Head Dam. The water at this location comes from two water sources, Butte Creek and water from the west branch of the Feather River. From July through September, the west branch of the Feather River provides approximately 40 percent of the flows downstream from the Centerville Head Dam in the anadromous reach of Butte Creek. The water from the Feather River is diverted at the Hendricks Head Dam and flows through the Hendricks/Toadtown Canal where it merges with Butte Creek water from the Butte Canal that is diverted at the Butte Head Dam. The water continues through the DeSabla Forebay, and then reconnects to Butte Creek. Water also flows through the mainstem of Butte Creek between the Butte Head Dam and the DeSabla Forebay confluence. The water is again diverted at the Centerville Head Dam, where a majority of the water is sent down the Centerville Canal, and reconnects to Butte Creek at the Centerville Powerhouse. PG&E is required to maintain a minimum flow of 40 cfs in the mainstem of Butte Creek between the Centerville Head Dam and the Centerville Powerhouse from June 1 through September 14. In recent years, PG&E has voluntarily increased the minimum flow to 60 cfs during the onset of spawning, in late September. PG&E also has a contingency plan for when air temperatures exceed 105 F (typically in the middle of the summer), in which they alter the flow regime to provide colder water to the reach where spring-run Chinook salmon are over-summering above the Centerville Powerhouse. Spring-Run Chinook Salmon 3-17 November 2010

32 San Joaquin River Restoration Program Figure 3-6. Reaches of Butte Creek and West Branch of the Feather River Controlled by Pacific Gas & Electric Company Affecting Butte Creek Spring-Run Chinook Salmon, Including Temperature and Flow Gage Locations and Distances 3-18 November 2010 Spring-Run Chinook Salmon

33 3.0 Stock Descriptions Water Temperature Water temperatures are regularly monitored seasonally from June through September throughout the PG&E DeSabla-Centerville Project. PG&E in consultation with DFG, NMFS, and USFWS, has developed a Project Operations and Management Plan that includes a contingency for extreme heat events (beginning in 2004). PG&E prepares weekly weather forecasts, based on USFS weather stations, for the DeSabla-Centerville Project Area, which encompasses the Butte Creek spring-run Chinook salmon s holding and spawning area. If air temperatures will exceed 105 F (41 C) for 2 or more days then, in consultation with the Resource Agencies, PG&E changes the flow regime by altering the flow amount and location of release to reduce the water temperatures within the DeSabla-Centerville Project Area. The water temperature in the holding and spawning habitat frequently exceeds 59 F (15 C) from July through September (Figure 3-7). PG&E is required to maintain a minimum flow of 40 cfs in Butte Creek between the Centerville Head Dam and the Centerville Powerhouse from June 1 through September. Since 2004, PG&E has voluntarily increased the minimum flow in Butte Creek to 60 cfs during the onset of spring-run Chinook salmon spawning. This increase has reduced water temperatures in this section of river and has increased the amount of usable spawning gravel by approximately 26 percent TEMPERATURE (C) /1 7/8 7/15 7/22 7/29 8/5 8/12 8/19 8/26 9/2 9/9 9/16 9/23 9/30 DATE Source: McReynolds et al Figure 3-7. Mean Daily Water Temperature ( C) at Quartz Bowl Pool for Period July Through September Spring-Run Chinook Salmon 3-19 November 2010

34 San Joaquin River Restoration Program Observed disease outbreaks within the Butte Creek spring-run Chinook salmon population have generally occurred during the summer holding period. In 2002, there were approximately 3,431 pre-spawn mortalities out of an estimated population of 16,328; in 2003 there were approximately 11,231 pre-spawn mortalities out of an estimate population of 17,294; and during 2004 there were approximately 418 pre-spawn mortalities out of an estimated population of 10,639 (Ward et al. 2007). In 2003, fish mortality was attributed to the high number of fish concentrated in limited holding pools with high water temperatures, and an outbreak of two diseases Flavobacterium columnare (Columnaris) and the protozoan Ichthyophthirius multiphilis (Ich) (Williams 2006). The mortalities during 2002 and 2003 coincided with significant daily average water temperatures above 67 F (19.5 C). The pre-spawn mortalities during 2004 were concluded to be the normal attrition for salmon holding in fresh water since early spring. During the 2004 summer months, the average air and water temperatures were generally lower than in 2002 and 2003, and Butte Creek flows were slightly higher. The pre-spawn mortalities in subsequent years (2005 through 2007) were also concluded to be due to normal attrition Life History/Phenotypic Expression Upstream Migration and Holding The entire available holding and spawning area for Butte Creek spring-run Chinook salmon is below 931 feet in elevation, due to a 15-foot waterfall barrier known as the Quartz Bowl Falls. The best holding and spawning habitat for the spring-run Chinook salmon is within approximately 11 miles of the river, from Quartz Pool downstream to the Centerville Covered Bridge (Ward et al. 2004). The highest quality and quantity of holding habitat is within the uppermost 3 miles (from Quartz Pool to Whiskey Flat). Another good holding location is directly below the Centerville Powerhouse, due to the cooler water found there. The diversion at the Centerville Head Dam, which sends water down the Centerville Canal to the Centerville Powerhouse, which significantly reduces water temperatures directly below the powerhouse due to reduced transition time and shading. Butte Creek spring-run Chinook salmon adults migrate from February through June, with the peak in mid-april. Adult migration is frequently impaired by low flows and high water temperatures in June, and adult Chinook salmon that have not migrated above State Highway 99 by mid-june have a lower likelihood of surviving to spawn. DFG biologists also regularly observe large numbers of spring-run Chinook salmon holding directly below the Centerville Powerhouse. During the 7-year period from 2001 to 2007, approximately 60 percent of the fish held above the Centerville Powerhouse and 40 percent held below it (Figure 3-8) (McReynolds and Garman 2008) November 2010 Spring-Run Chinook Salmon

35 3.0 Stock Descriptions ABOVE POWERHOUSE AVG. HOLDING: 8,063 (60%) MAX. HOLDING: 12,608 MIN. HOLDING: 2,531 EST. SPAWNING CAPACITY AT 45CFS : 416-1,431 EST SPAWNING CAPACITY AT 80 CFS : 629-2,164 BELOW POWERHOUSE AVG. HOLDING: 5,306 (40%) MAX. HOLDING: 9,281 MIN. HOLDING: 2,014 EST. SPAWNING CAPACITY AT 80 CFS: 2,428-8,352 EST. SPAWNING CAPACITY AT 120 CFS: 2,914-10, NUMBER HOLDING REACH 2007 (N = 6,858) 2006 (N = 6,547) 2005 (N = 17,615) 2004 (N = 10,639) 2003 (N = 17,294) 2002 (N = 16,327) 2001 (N = 18,312) Source: McReynolds et al Figure 3-8. Distribution by Reach of the Number of Butte Creek SRCS Holding, During Spring-Run Chinook Salmon 3-21 November 2010

36 San Joaquin River Restoration Program Spawning The highest quality and quantity spawning gravel is within the first 5 miles directly below the Centerville Powerhouse. Estimates of available spawning habitat based on maximum suitable flows (130 cfs) concluded that approximately 18 percent of the suitable spawning gravel is located above the Centerville Powerhouse and 82 percent below (Ward et al. 2004). The maximum number of spawners at these locations is 152 to 1,316 at 40 cfs above Centerville Powerhouse, and 270 to 2,352 at 40 cfs and 1,262 to 10,976 at 130 cfs below (Ward et al. 2004). The spring-run Chinook salmon generally spawn between late-september through early November, with the peak in early October. During the 7-year period from 2001 to 2007, approximately 45 percent of the fish spawned above the Centerville Powerhouse and 55 percent below (Figure 3-9) (McReynolds et al. 2008). During 2004, PG&E increased the flow above the Centerville Powerhouse from 40 cfs to 60 cfs to provide additional habitat for the spawning spring-run Chinook salmon. The increase in flow increased the amount of usable spawning gravel by approximately 26 percent (Ward et al. 2004) ABOVE POWERHOUSE AVG. SPAWNING : 5,054 (46%) MAX. SPAWNING: 10,887 MIN. SPAWNING : 1,527 EST. SPAWNING CAPACITY AT 45CFS : 416-1,431 EST SPAWNING CAPACITY AT 80 CFS : 629-2,164 BELOW POWERHOUSE AVG. SPAWNING: 5,947 (54%) MAX. SPAWNING: 11,089 MIN. SPAWNING : 3,446 EST. SPAWNING CAPACITY AT 80 CFS: 2,428-8,352 EST. SPAWNING CAPACITY AT 120 CFS: 2,914-10, NUMBER SPAWNING REACH 2007 (N = 6,220) 2006 ( N = 6,303) 2005 (N = 16,998) 2004 (N = 10,221) 2003 (N = 6,063) 2002 (N = 12,897) 2001 (N = 18,312) Source: McReynolds et al Figure 3-9. Distribution by Reach of the Number of Butte Creek SRCS Spawning, During November 2010 Spring-Run Chinook Salmon

37 3.0 Stock Descriptions Outmigration Butte Creek spring-run Chinook salmon generally outmigrate as fry from November through February, and rear below the Parrott-Phelan Diversion Dam. The outmigration movements are heavily influenced by flow. Most spring-run Chinook salmon rear in the Sutter Bypass from February through May, and then migrate into the Sacramento River and continue to the Sacramento-San Joaquin Delta (Delta). Some fish will rear above the Parrott-Phelan Diversion Dam, in the mainstem of Butte Creek. These fish will generally rear for 12 or more months before outmigrating. Rearing The highest quality and quantity of juvenile rearing habitat is located in the Sutter Bypass, due to the connection to the floodplain (Williams 2006). Butte Creek spring-run Chinook salmon generally rear in the Sutter Bypass. Floodplain productivity increases with spring temperatures and residence times provide advantageous resources for outmigrating juveniles. Juvenile Chinook salmon that rear in the floodplain have significantly higher growth rates than fish that rear in riverine habitats (Moyle et al. 2008). In fact, spring-run Chinook salmon were captured and tagged at the Parrott- Phelam Diversion Dam and recaptured in the Sutter Bypass. DFG biologists have calculated the average growth rate of juvenile spring-run Chinook salmon for the Sutter Bypass recaptures to be 0.52 millimeter (mm)/day during 1999, 0.66 mm/day during 2000, and 0.38 mm/day during 2002 (Ward and McReynolds 2004). Every year there are generally a handful of yearlings observed during spring-run Chinook salmon surveys. These salmon rear above Parrott-Phelan Diversion Dam, in the mainstem of Butte Creek. These fish grow to approximately 150 mm FL and remain in Butte Creek above the Parrott-Phelan Diversion Dam for 12 months or more before leaving Butte Creek and outmigrating to the Delta as yearlings (Ward et al. 2004b) Population Size The data below is based on DFG escapement estimates for the years 1954 through 2006 (Table 3-4). The approximate averages for the last 30, 20, and 10 years are 3,000, 4,400, and 7,400, respectively. Spring-Run Chinook Salmon 3-23 November 2010

38 San Joaquin River Restoration Program Table 3-4. Butte Creek Spring-Run Chinook Salmon Spawning Escapement Estimates for the Period 1954 Through 2008 Year Run Size Year Run Size Year Run Size Year ,679* ,118* Run Size , Snorkel Prespawn Mortality Spawn , , ,312** , , ,785 3,431 12, ,300* ,398 11,231 6, , * , , , * , , , * , , , * , , * , , ,500* ,413* * ,212* Source: McReynolds 2008 and DFG 2009 Notes: * Surveys before 1989 used various methods with varying precision. Snorkel surveys implemented since 1989 are thought to significantly underestimate the actual population size and should only be used as an index. Spawning surveys results for were generated by a modified Schaefer Model carcass survey. ** Number as reported for 2001 (22,744) in error (Ward et al. 2004). During a 7-year period from 2001 to 2007, the average size of females was 762 mm and the average size of males was 793 mm. The average size of both males and females were significantly higher in 2007, 2006, and 2003, with males averaging 833 mm and females averaging 775 mm, compared with 2001, 2002, 2004, and 2005, with males averaging 762 mm and females averaging 711 mm. This size distribution is likely due to the percentage of different age classes. Spring-run Chinook salmon generally return at Age 3 or Age 4, and the compositions of the two age classes vary each year. Between 2001 and 2007, Age 4 dominated the adult composition in Years 2006 and 2003, 75 percent and 69 percent, respectively. Whereas in the 2001, 2002, 2004, and 2005, the adult composition was dominantly Age 3, 89 percent, 86 percent, 89 percent, and 97.5 percent. In 2007, the adult composition was approximately evenly distributed, 53 percent of the population was Age 3 and 47 percent was Age November 2010 Spring-Run Chinook Salmon

39 3.0 Stock Descriptions Hatchery Influence There is little hatchery influence on the Butte Creek spring-run Chinook salmon population. No hatcheries exist on Butte Creek and the stream has not historically and is not currently planted with hatchery fish. The only exception was in 1986, when 200,000 juvenile Feather River spring-run Chinook salmon hatchery fish were planted into Butte Creek due to the extreme low levels of returns of Butte Creek spring-run Chinook salmon (Moyle et al. 2008). However, it is not believed that this plant had any genetic effect on the Butte Creek population (Garza et al. 2008). Hatchery Chinook salmon occasionally stray into Butte Creek, but in very low numbers. 3.4 Other Central Valley Phenotypic Spring-Run Chinook Salmon Populations In addition to the recognized stocks listed above, evidence exists in other Central Valley watersheds of the occurrence of Chinook salmon displaying the spring-run Chinook salmon phenotype. These small localized occurrences warrant consideration because they occur in watersheds in closer proximity to the San Joaquin River geographically, and thus may be more adapted to the local conditions that will occur in the San Joaquin River. Two such watersheds in which data exists on phenotypic spring-run Chinook salmon are the Mokelumne River, an eastside tributary to the Delta, and the Stanislaus River, a tributary to the San Joaquin River Existing Conditions Mokelumne River The lower Mokelumne River is considered an Eastside Tributary to the Delta. Its confluence with the San Joaquin River is within the Delta proper boundaries. Flows in the Mokelumne River are regulated by a Joint Settlement Agreement (JSA) (1998) under Federal Energy Regulatory Commission (FERC) license. As such, the Mokelumne flow is based on water year types derived from precipitation, snow pack, and available storage in Camanche and Pardee reservoirs. Flow varies for the five water year types; wet, normal and above, below normal, dry, and critically dry. Minimum flow schedules are based on fall-run Chinook salmon life history and separated into fall (migration/spawning flows), winter (incubation flows), spring (emigration flows), and summer base flows. Minimum summer base flows range from 80 cfs in wet years to 20 cfs in dry and critically dry. Few holding pools are available for over-summering spring-run Chinook salmon on the Mokelumne, and summer temperatures typically reach 64 F (18 o C) Camanche Dam is on River Kilometer (RKM) 103 and is the upper limit to anadromy on the Mokelumne River (Figure 3-10). Camanche Dam blocks approximately 80 percent of historical Chinook spawning habitat (DFG 1991). There are approximately 16 km of spawning habitat downstream from Camanche Dam available for salmonid spawning, and holding habitat is limited to a few large pools in the first river mile below Camanche Dam. Spring-Run Chinook Salmon 3-25 November 2010

40 San Joaquin River Restoration Program Source: East Bay Municipal Utility District Figure Mokelumne River 3-26 November 2010 Spring-Run Chinook Salmon

41 3.0 Stock Descriptions Stanislaus River The Stanislaus River is one of three major tributaries to the San Joaquin River (Figure 3-11). It is snow fed and its headwaters begin at an elevation of approximately 3,675 m. Like all San Joaquin River tributaries, multiple dams are located on the upper Stanislaus River. Historically, various life history types of Chinook salmon inhabited the Stanislaus River, including fall-, late fall-, and spring-runs (Reynolds et al. 1993). Currently, upstream migration for anadromous fishes ends at Goodwin Dam, RKM 94. Historically, upstream migration and spawning occurred well into the Stanislaus River s three forks, but miles of spawning and rearing habitat were lost due to dam construction (Fry 1961). Source: Anderson et al Figure Stanislaus River Spring-Run Chinook Salmon 3-27 November 2010

42 San Joaquin River Restoration Program Life History/Phenotypic Expression Mokelumne River Year-round video monitoring on the Mokelumne River began in Since that time, it has become clear that adult Chinook salmon are ascending the Mokelumne River from April through June on an irregular basis, in addition to the well-established population of fall-run Chinook salmon (escapement from August/September through January). Migration. Phenotypic spring-run Chinook salmon observed on the Mokelumne River have passed video monitoring between April and June in low numbers. Holding and Spawning. Limited holding opportunities exist on the Mokelumne River. There are few large pools in the uppermost reach just below Camanche Dam. No assessments of holding or spawning have been conducted. Rearing. No assessment of spring-run Chinook salmon rearing has been conducted due to the confounding effects of spatial and temporal overlap with fall-run Chinook salmon, and the relatively small population size of phenotypic spring-run Chinook salmon spawners. Outmigration. Yearling-sized juvenile Chinook (more than 100 mm FL) have been observed in rotary screw trapping in low numbers in December and January of some years (Workman 2006a, Workman 2002a). Rotary screw traps are typically installed in mid-december and operated until June or July, depending on water year type. Stanislaus River In 2002, a resistance board weir was installed on the Stanislaus River to assess escapement numbers and timing of Chinook salmon and steelhead trout (Oncorhynchus mykiss). In 2003, the weir was improved with the addition of an infrared camera. Migration. Phenotypic spring-run Chinook salmon have been observed passing the weir on the Stanislaus River in May and June (Anderson et al. 2007). Holding and Spawning. Chinook salmon have been reported in the Stanislaus River during the summer months. Snorkel surveys (Kennedy and Cannon 2005) conducted between October 2002 and October 2004 identified adults in June 2003 and June 2004 between Goodwin and Lovers Lead. Snorkel surveys also observed Chinook fry in December 2003 at Goodwin Dam, Two Mile Bar, and Knights Ferry, which indicates spawning occurring in September. In 2000, DFG (unpublished data) seined a deep pool at Bottonbush Recreation Area on five occasions, between June 29 and August 25, and captured 28 fish. Of these, eight were adipose fin-clipped and five had coded wire tags (CWT). All CWT fish originated from the FRFH. Rearing. No assessment of spring-run Chinook salmon rearing has been conducted due to the confounding effects of spatial and temporal overlap with fall-run Chinook salmon, and the relatively small population size of phenotypic spring-run Chinook salmon spawners November 2010 Spring-Run Chinook Salmon

43 3.0 Stock Descriptions Outmigration. Rotary screw traps have captured low numbers of yearling smolts (defined as more than110 mm) on the Stanislaus River from February to April (Watry et al. 2007) Existing Population Size Mokelumne River Phenotypic spring-run Chinook salmon on the Mokelumne River have numbered as high as 114 in the spring of 2002 between April and July, with four adipose fin clipped fish (i.e., hatchery origin fish) observed (Workman 2002b). Ninety-seven were observed in 2003 between March and July, with 21 adipose fin clipped fish observed (Workman 2003). None were observed in 2004, and in 2005, 2006, and 2007, limitations in video monitoring due to construction led to carcass survey data for escapement estimates, and no estimate of phenotypic spring-run Chinook salmon were attempted (Workman 2004, 2005, 2006b, Workman and Rible 2007, Workman et al. 2008). Stanislaus River In 2007, 11 phenotypic spring-run Chinook salmon were observed passing the weir between May and June. Future monitoring will determine if these fish are a typical occurrence or an anomaly (Anderson et al. 2007) Hatchery Influence and Interbasin Transfers Mokelumne River The Mokelumne River has a DFG fall-run Chinook salmon production hatchery at the base of Camanche Dam. Historically, the hatchery has imported eggs and fry from both the Nimbus Fish Hatchery and the FRFH to meet production goals. Stanislaus River There is no hatchery on the Stanislaus River. Hatchery stock, identified by adipose fin clips, have been detected during weir operations denoting a small portion of hatchery influence is occurring in the watershed (Anderson et al. 2007). During carcass surveys in 2009, 11 percent of Chinook adults were adipose fin clipped (DFG unpublished data) Genetics Genetics work on these populations to determine if they are spring-run Chinook salmon has not been conducted, and although these populations exhibit the spring-run Chinook salmon phenotype, genetic analysis needs to be conducted to determine whether these fish are genetically or just phenotypically spring-run Chinook salmon. Spring-Run Chinook Salmon 3-29 November 2010

44 San Joaquin River Restoration Program This page left blank intentionally November 2010 Spring-Run Chinook Salmon

45 4.0 Population Genetics There are only three stocks of spring-run Chinook salmon ESU Chinook salmon in the Central Valley that are possible donors for the reintroduction project in the San Joaquin River. These are the Butte Creek stock, the Mill Creek/Deer Creek stock, and the Feather River stock. Banks et al. (2000) and Garza et al. (2008) have shown that these three stocks are genetically distinct, and that the Mill Creek and Deer Creek populations are essentially the same stock. There are additional small populations of spring-run Chinook salmon in the Central Valley (e.g., Big Chico, Antelope, and Clear creeks, Mokelumne River, and Stanislaus River), but none of these, other than that on the Yuba River (Garza, unpublished data), have been confirmed to be from the Central Valley spring-run Chinook salmon ESU genetic lineage and may be early returning fall-run Chinook salmon. Even if these small populations were of Central Valley spring-run Chinook salmon ESU stocks, these runs are not appropriate as sole donor stocks for the SJRRP because they are too small and inconsistent to provide adequate numbers and diversity on which to base reintroduction. Only the three stocks mentioned above were, therefore, carefully evaluated as a potential primary or sole donor stock for the San Joaquin River reintroduction project. The three remaining spring-run Chinook salmon lineages are all in the northern part of the Central Valley in the Sacramento River subbasin. The San Joaquin River subbasin has, unfortunately, either completely or almost completely lost its spring-run Chinook salmon populations, although there are persistent reports of a small number of spring-run Chinook salmon returning to the Mokelumne and Stanislaus rivers (Workman and Merz, field observations). The Deer/Mill Creek population is the northernmost of these and therefore the furthest from the San Joaquin River, with the Butte Creek population just to the south and the Feather River the geographically most proximate of these three potential donor stocks. The Deer/Mill Creek population also has the lowest current abundance, with escapement estimates of about 3,389, 1,564, and 502 in 2006, 2007, and 2008, respectively. Butte Creek has a larger census population size, with current escapement estimates of 4,579, 4,943 and 3,935 in 2006, 2007, and 2008, respectively. However, these escapement estimates use different methodology (carcass counts vs. snorkel survey), so they are not directly comparable, and the Butte Creek estimates are likely more comprehensive than those for Deer/Mill Creek. Furthermore, it is important to note that, over the last 20 to 30 years the mean census size estimates of the two stocks have been similar, and both historical and current population sizes are important in determining levels of genetic variation. Escapement estimates for the Feather River spring-run Chinook salmon populations are not complete, since the Feather River stock escapement estimates use a different methodology and only attempt to enumerate hatchery fish. The escapement estimates for the hatchery component only in 2006, 2007, and 2008, were 2,061, 2,674 and 1,418 fish, Spring-Run Chinook Salmon 4-1 November 2010

46 San Joaquin River Restoration Program respectively. Since the non-counted, naturally spawning component of this stock is typically large, the census size of the Feather River stock is likely the largest of the three spring-run Chinook salmon stocks (DFG 2009). There are three datasets available to evaluate the relative genetic diversity of the three potential spring-run Chinook salmon donor stocks for the San Joaquin River reintroduction project. The first of these is published in Banks et al. (2000) and consists of microsatellite data for the Deer/Mill Creek and Butte Creek stocks. While a substantial number of fish were sampled for this study, this dataset unfortunately does not include fish from the Feather River spring-run Chinook salmon stock. It also includes data from only a small number of microsatellite loci, with an average of only about seven loci per fish genotyped. As such, the two primary measures of genetic diversity are significantly affected by sampling variance. The first measure, observed heterozygosity, is essentially identical in the two stocks 0.61 vs in the Deer/Mill and Butte Creek stocks, respectively). Allelic diversity, as measured by the average number of alleles observed per locus, is about 7 percent higher in the Deer/Mill Creek stock than in the Butte Creek stock (6.60 vs respectively). It is worth noting that, for microsatellite loci, the number of alleles is a more sensitive indicator of recent effective population size than heterozygosity (Garza and Williamson 2001), so these data are indicative of higher effective population size and consequent greater genetic diversity in Deer/Mill Creek than in Butte Creek spring-run Chinook salmon. The second dataset available for evaluation is that of Garza et al. (2008) and consists of data for 20 microsatellite loci from Chinook salmon sampled in 2002 and 2003 throughout the Central Valley, including all three of the known, extant spring-run Chinook salmon ESU stocks. In this analysis, the Deer/Mill Creek spring-run Chinook salmon populations were considered separately and differences in the sample sizes for the different spring-run Chinook salmon stocks necessitated the use of allelic richness, a measure of the number of alleles that takes into account such differences (Petit et al. 1998). With this large microsatellite dataset, the mean allelic richness per locus of the Mill Creek, Deer Creek, Butte Creek, and Feather River stocks were 11.09, 10.85, 9.76 and 11.25, respectively. The observed heterozygosities were 0.77, 0.77, 0.74 and 0.78, respectively. It is worth noting that, aside from the Sacramento River winter-run Chinook salmon, the Butte Creek spring-run Chinook salmon stock had the lowest values of these two measures of genetic diversity of any Central Valley (or Klamath River) salmon population examined. It is also worth noting that, the Feather River spring-run Chinook salmon stock has been affected by hybridization with fall-run Chinook salmon, and at least some of the additional genetic diversity seen is likely due to the addition of fall-run Chinook salmon genes (Garza et al. 2008). The third dataset consists of recent unpublished data from 169 SNP loci. These SNP loci were developed by the Genetic Analysis of Pacific Salmonids (GAPS) consortium and by the Molecular Ecology and Genetic Analysis Team of the Southwest Fisheries Science Center, (Garza unpublished). These loci were developed with the dual objectives of developing intergenerational genetic tags for parentage-based tagging (PBT) and as markers for genetic stock identification (GSI) in fishery and ecological investigations. 4-2 November 2010 Spring-Run Chinook Salmon

47 4.0 Population Genetics For these 169 SNP loci, data were available for the Deer/Mill Creek (N=71), Butte Creek (N=54), and Feather River (N=94) spring-run Chinook salmon stocks. Since SNP loci generally only have two alleles, smaller numbers of fish are necessary to estimate perlocus measures of genetic diversity for SNPs than for microsatellite loci. However, these SNP loci were discovered using a panel of fish that included Central Valley spring-run Chinook salmon, so ascertainment bias will affect measures of allelic diversity and they are expected to be less informative than the corresponding measures for microsatellites. This is because they represent the proportion of polymorphic loci, with the mean number of alleles equals two when all loci are polymorphic and equals one when all loci are monomorphic, but only SNPs that were variable in the Central Valley were included in this set of genetic markers. The SNP dataset found similar measures of the mean number of alleles, with 1.91, 1.88 and 1.91 in the Deer/Mill Creek, Butte Creek, and Feather River stocks, respectively. Observed heterozygosity was more variable, with values of 0.29, 0.26 and 0.31 in the Deer/Mill Creek, Butte Creek, and Feather River stocks, respectively. In summary, all of the measures of genetic diversity in all of the datasets were the lowest for Butte Creek, intermediate for Deer/Mill Creek, and the highest for Feather River spring-run Chinook salmon. The effective population size of the Butte Creek spring-run Chinook salmon is, therefore, also the smallest of the three, since effective size determines the amount of genetic variation that is maintained in a population. The Butte Creek spring-run Chinook salmon stock then also has the highest risk of inbreeding in a reintroduction project. In contrast, the Feather River spring-run Chinook salmon stock has the highest genetic diversity of the three. However, this stock is known to have been affected by hybridization with fall-run Chinook salmon at the FRFH (Garza et al. 2008) and hybridization is ongoing (Kindopp pers. comm.). It is also likely that hybridization occurs in the spawning grounds of the lower Feather River. At least some of the additional genetic diversity seen in the Feather River stock is likely due to the addition of fall-run Chinook salmon genes. The Feather River spring-run Chinook salmon population is more genetically similar to fall-run Chinook salmon in the Feather River than to the spring-run Chinook salmon in the Deer/Mill Creek and Butte Creek populations, raising the potential for outbreeding depression during an introduction. This is unfavorable for the maintenance of phenotypic differentiation (i.e., spring-run Chinook salmon offspring returning as fall-run Chinook salmon); however, it also reduces the risk of inbreeding in a reintroduction project and the consequent reduction in fitness from inbreeding depression. Conversely, tagging studies have found that some offspring from Feather River spring-run Chinook salmon return as fall-run Chinook salmon, and vice versa (DFG 1998) Another aspect of the genetic/demographic history of the three spring-run Chinook salmon stocks that needs to be considered is the relative influence of hatchery-produced fish on the naturally spawning stock. The FRFH, which began operation in 1967, has produced and released millions of juvenile salmon, both spring- and fall-run Chinook salmon, annually for more than 40 years. These fish have extensively introgressed with naturally spawning populations in the Feather River and elsewhere. In contrast, the Deer/Mill Creek and Butte Creek spring-run Chinook salmon stocks appear to be largely free of introgression from hatchery-produced fish. There is accumulating evidence that Spring-Run Chinook Salmon 4-3 November 2010

48 San Joaquin River Restoration Program salmon from hatchery stocks are less fit than natural origin fish (Berejikian and Ford 2004, Myers et al. 2004), and that this is at least partly due to hatchery domestication selection, which often causes maladaptation to environmental conditions in natural areas. However, domestication selection from hatchery fish can be counteracted relatively quickly by crossing with natural origin fish and subsequent selection in natural areas (Quinn et al. 2000, Unwin et al. 2000), as long as the artificial selection is not coincident with a loss of genetic variation and an increase in inbreeding. 4-4 November 2010 Spring-Run Chinook Salmon

49 5.0 Lower San Joaquin River Existing Conditions The Restoration Area, approximately 153 miles long, extends from Friant Dam at the upstream end near the town of Friant, downstream to the confluence of the Merced River, and includes an extensive flood control bypass system (Figure 5-1). The Restoration Area has been significantly altered by changes in land and water use over the past century. Five river reaches have been defined to address the great variation in river characteristics throughout the Restoration Area. The reaches are differentiated by their geomorphology and resulting channel morphology, and by the infrastructure along the river. Hence, flow characteristics, geomorphology, and channel morphology are similar within each of the reaches. The characteristics of these Reaches are described in further detail in Chapter 2 of the FMP. Spring-Run Chinook Salmon 5-1 November 2010

50 San Joaquin River Restoration Program Figure 5-1. San Joaquin River Restoration Area and the Defined River Reaches 5-2 November 2010 Spring-Run Chinook Salmon